Download - First Sliding Mode Control of a DFIG-based Wind Turbine ...source of energy. The performance of a wind turbine depends essentially on three parameters: the power of the wind, the turbine

First Sliding Mode Control of a DFIG-based Wind Turbine with

mechanical part control

LARBI DJILALI1, OSCAR J. SUAREZ1, ALDO PARDO GARCIA2, MOHAMMED BELKHEIRI3

Electrical Engineering Department 1CINVESTAV, 2University of Pamplona, 3University Amar Telidji Av. del Bosque 1145-Col. El Bajío, Campus Pamplona, Laghouat

MEXICO, COLOMBIA, ALGERIA [email protected], [email protected], [email protected], [email protected]

Abstract: - The objective of this article is the study of the dynamic behavior of a wind turbine based on doubly fed induction generator (DFIG) with Maximum Power Point Tracking (MPPT) without enslavement of speed and PITCH control. The DFIG stator is directly connected to the grid however; the rotor is linked via a power electronic converter (rectifier, filter, and inverter). The stator active and reactive powers are controlled directly by adjusting the DFIG rotor quantities, while applying direct vector control based on a first order sliding mode controller (SMC).

Key-Words: - Wind turbine, MPPT, PITCH, DFIG, active and reactive power, sliding mode. 1 Introduction

Nowadays, wind energy has many advantages, because it does not pollute and is an inexhaustible source of energy. The performance of a wind turbine depends essentially on three parameters: the power of the wind, the turbine power curve, and the ability of the generator to respond to fluctuations in the wind. The DFIG is a machine which has shown an excellent performance and is widely used in the wind turbine industry [1]. There are many reasons to use a DFIG for variable speed wind turbine, such as reducing stress on the mechanical parts, noise reduction, and the possibility of the active and reactive powers control [1].

A wind system based on DFIG using a ‘back-to-back’ converter to connect the rotor of the generator with the network presents many advantages; one is being that the power converters are sized to move a fraction of the total power of the system [2]. To maximize this energy, a wind turbine should include [3] [4]:

A system that allows mechanical control A system that allows electrical control

In this article, we present a sliding mode control strategy applied to control; the MPPT algorithm without enslavement of the rotation speed, and PITCH system to control aero-generator. In section 2 the wind turbine modeling is presented. Then, in section 3, the strategy of the command of the

mechanical part, including the MPPT algorithm without enslavement of rotational speed and the blade orientation (PITCH) system is studied. In the section 4, the model of the DFIG in the coordinate system (d,q) and vector control based on the SMC of the active and reactive DFIG powers is introduced. We conclude with the presentation of the simulation results.

2 Modeling of the wind turbine 2.1 Modeling of the turbine The theoretically undisturbed power of the wind through a surface S without loss of speed is [4]:

𝑷𝒘𝒊𝒏𝒅 =𝟏

𝟐𝝆𝑺𝒗𝟏

𝟑 (1)

where 𝑷𝒘𝒊𝒏𝒅 is the wind power (W) and, ρ is the air volume density (Kg/m3). Fig 1 presents the horizontal axis wind system where 𝒗𝟏 and 𝒗𝟐 are upstream and downstream of wind speed, respectively. The tube surface area before and after are given by 𝒔𝟏, 𝒔𝟐 and the swept area of turbine blades is given by S. The equation which expresses the incompressibility of the air and the permanence of flow is written by

𝑷𝒕𝒖𝒓𝒃 =𝟏

𝟐𝝆𝑺𝒗(𝒗𝟏

𝟐 − 𝒗𝟐𝟐) (2)

WSEAS TRANSACTIONS on SYSTEMSLarbi Djilali, Oscar J. Suarez,

Aldo Pardo Garcia, Mohammed Belkheiri

E-ISSN: 2224-2678 71 Volume 17, 2018

mailto:[email protected]



Figure 1. Stream around a wind turbine tube where 𝑷𝒕𝒖𝒓𝒃 is the power extracted by the turbine. When the speed of the air passing through the turbine is given by [5][6],

𝒗 =𝟏

𝟐(𝒗𝟏 + 𝒗𝟐) (3)

then the ratio between the extracted power of the wind and the total power theoretically available is

𝑪𝒑 =𝑷𝒕𝒖𝒓𝒃

𝑷𝒘𝒊𝒏𝒅=

𝟏

𝟐|𝟏 − (

𝒗𝟏

𝒗𝟐)𝟐||𝟏 + (

𝒗𝟏

𝒗𝟐)|

then, 𝑷𝒕𝒖𝒓𝒃 = 𝑪𝒑𝑷𝒘𝒊𝒏𝒅 =

𝟏

𝟐𝑪𝒑𝝆𝑺𝒗𝟑 (4)

This is exactly the power absorbed by the wind turbine. From the definition of the relative speed, λ can be found as follows:

𝝀 =𝜴𝒕𝒓

𝒗 (5)

where 𝜴𝒕 is the turbine’s mechanical speed, 𝒓 is its radius, and 𝒗 is wind speed. The power coefficient 𝑪𝒑 represents the aerodynamic performance of the wind turbine and also depends on the characteristic of the turbine. This factor presents a theoretical limit, called the BETZ limit, which is equal to 0.593 and is not achievable in practice [7]. The study of a particular turbine allowed deducing the following empirical formula for variable speed wind turbines: 𝑪𝒑 = 𝑪𝟏 (𝑪𝟐

𝟏

𝜦− 𝑪𝟑𝜷𝒙 − 𝑪𝟒 − 𝑪𝟓) 𝒆𝒙𝒑 (

−𝑪𝟔

𝜦) (6)

where, 𝑪𝟏 = 𝟎. 𝟓; 𝑪𝟐 =116; 𝑪𝟑 = 0.4; 𝑪𝟒 =0.4; 𝑪𝟓 = 𝟓 ;

𝑪𝟔 = 𝟐𝟏; 𝟏

𝜦=

𝟏

𝝀+𝟎.𝟎𝟖𝜷−

𝟎.𝟎𝟑𝟓

𝟏+𝜷𝟑

Fig. 2 shows the curves of the power factor 𝝀 for different values of blade pitch angle 𝜷.We obtain a coefficient of maximum power of 0.3522 for a speed ratio 𝝀 equal to 9.7 (𝝀𝒐𝒑𝒕).

Figure 2. Power based on λ coefficient and β

angle

2.2 The multiplier mode

The multiplier is the link between the turbine and the generator. It is assumed to be rigid and modeled by a simple gain [8] [9]:

𝑻𝒎𝒆𝒄 =𝟏

𝑮𝑻𝒕 (7)

𝜴 = 𝑮𝜴𝒕 (8)

where 𝑻𝒎𝒆𝒄 is mechanical torque (N.m), 𝑻𝒕is mechanical torque (N.m) on the axis of the turbine, 𝑮 is the multiplier gain, 𝜴 is the Mechanical speed (rad/s), and 𝜴𝒕 is the Turbine mechanical speed (rad/s). We can write the fundamental equation of the mechanical system dynamics of the DFIG as follows:

𝑱𝒅𝜴

𝒅𝒕= 𝑻𝒎𝒆𝒄 − 𝑻𝒆𝒎 − 𝒇𝜴 (9)

where 𝐉 is the total inertia of system (Kg.m2), 𝑻𝒆𝒎is electromechanical torque and, 𝐟 is the coefficient of friction (N.m/s). These equations provide the turbine block diagram in Fig. 3.

Figure 3. Aerodynamic and mechanical wind

turbine model

0 5 10 15-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

X: 9.701

Y: 0.3556

Relative speed

Pow

er

coeffic

ient C

p

= 20

= 40

= 60

= 80

= 100



E-ISSN: 2224-2678 72 Volume 17, 2018

3 Control of the Mechanical Part 3.1 The MPPT algorithm The operation of a wind turbine at a variable speed can be defined according to three areas [10] [11], as shown in Fig. 4. In zone 1, the wind speed is low, insufficient to start the turbine. In zone 2, the wind reaches 𝒗𝒎𝒊𝒏 , allowing startup. The wind turbine will operate so as to extract the maximum available power MPPT algorithm for optimal operation until the wind reaches 𝒗𝒏. In zone 3, the wind reaches speeds higher than the rated speed; the rotational speed and mechanical power will be maintained at their nominal values so as not to damage the wind turbine. The speed limitation of the aero-generator and the mechanical power generated by the turbine are assured by the PITCH system [11] [12].

Figure 4. Zones of operation of a wind turbine at

variable speed

In practice, a precise measurement of wind speed is difficult to achieve. Therefore, most wind turbines are controlled without enslavement of speed [13]. This command structure is based on the assumption that the wind speed varies very little in a permanent system and based on the following equation:

𝑻𝒆𝒎𝒓𝒆𝒇

=𝑪𝒑𝒎𝒂𝒙𝝆𝝅𝒓𝟓

𝟐𝝀𝒐𝒑𝒕𝟑𝑮𝟑

∗ 𝜴𝟐 (10)

The structure is shown in Fig. 5.

Figure 5. MPPT: control without enslavement of

the rotational speed

3.2 Modeling of the PITCH system Nowadays, high-powered wind turbines use the PITCH system for power control in zone 3. For this purpose, they use the principle of aerodynamic control to limit the extracted power to its nominal value [5]. The PITCH system control reduces the yield when β increases. Fig. 6 shows the different parts of an angle control system.

Figure 6. Diagram of the angle control block

4 Modeling and Control of the DFIG

4.1 Modeling of the DFIG The DFIG can be presented by means of different mathematical models according to the chosen variable states; the model proposed in this manuscript is given by: 𝐗 = [𝐈𝐬𝐝 𝐈𝐬𝐪 𝚽𝐫𝐝 𝚽𝐫𝐪 ]

𝐓 (11) Therefore, the DFIG model is given by:

��𝐬𝐝 = −𝛄𝐈𝐬𝐝 + 𝐰𝐬𝐈𝐪𝐬 +𝐤

𝐓𝐫

𝐫𝐝

+ 𝐤(𝐰𝐬 − 𝐰𝐫)𝐫𝐪

− 𝐤𝐕𝐫𝐝 +𝐕𝐬𝐝

𝛅𝐋𝐬

��𝐬𝐪 = −𝐰𝐬𝐈𝐪𝐬 − 𝛄𝐈𝐬𝐪 + 𝐤(𝐰𝐫 − 𝐰𝐬)𝐫𝐝

−𝐤

𝐓𝐫

𝐫𝐪

− 𝐤𝐕𝐫𝐪 +𝐕𝐬𝐪

𝛅𝐋𝐬

𝐫𝐝=

𝐋𝐦

𝐋𝐫

𝐈𝐬𝐝 −𝟏

𝐓𝐫

𝐫𝐝

+ 𝐰𝐫 𝐫𝐪+ 𝐕𝐫𝐝

𝐫𝐪=

𝐋𝐦

𝐋𝐫

𝐈𝐬𝐪 −𝟏

𝐓𝐫

𝐫𝐪

− 𝐰𝐫 𝐫𝐝+ 𝐕𝐫𝐪

with:

𝛄 = [𝟏

𝛅𝐓𝐬+

𝐋𝐦𝟐

𝛅𝐋𝐬𝐋𝐫𝐓𝐫] ; 𝐓𝐫 =

𝐋𝐫

𝐑𝐫 ;𝐓𝐬 =

𝐋𝐬

𝐑𝐬

𝐊 =𝐋𝐦

𝛅𝐋𝐬𝐋𝐫 ; 𝛅 = [𝟏 −

𝐋𝐦𝟐

𝐋𝐫𝐋𝐒]

where 𝑰𝒔𝒅, 𝑰𝒔𝒒, 𝒓𝒅, and 𝒓𝒒are the currents and flux of the stator and rotor; 𝑽𝒔𝒅, 𝑽𝒔𝒒, 𝑽𝒓𝒅 and 𝑽𝒓𝒒 are the voltages of the stator and rotor; 𝒘𝒔 and 𝒘𝒓 are the angular velocities of rotation; 𝑳𝒔 and 𝑳𝒓 are cyclic



E-ISSN: 2224-2678 73 Volume 17, 2018

inductances of a stator and rotor phase, respectively; and 𝑳𝒎 is the maximum mutual inductance between a stator and rotor phase.

The expression of electromagnetic torque (𝑻𝒆𝒎) of the DFIG is written as follows: 𝑻𝒆𝒎 = −

𝒑𝑳𝒎

𝑳𝒓(𝑰𝒔𝒒 𝒓𝒅 − 𝑰𝒔𝒅 𝒓𝒒) (12)

The active and reactive power stator of the DFIG is

{𝑷𝒔 = 𝑽𝒔𝒅𝑰𝒔𝒅 + 𝑽𝒔𝒒𝑰𝒔𝒒

𝑸𝒔 = 𝑽𝒔𝒒𝑰𝒔𝒅 − 𝑽𝒔𝒅𝑰𝒔𝒒

According to the model of active and reactive

power, we can see that the strong coupling between two power axes and the electromagnetic torque is the cross-product between stator flux and rotor currents, which makes it particularly difficult to control DFIG. In order to simplify the control of DFIG, we approximate its model to the continuous machine model, applying Filed-orient Control (FOC). In the FOC model, the DFIG has natural decoupling between two control axes; to apply the proposed method, we take:

𝒓𝒅 = 𝒔 and 𝒔𝒅 = 𝟎

So, the electromagnetic torque is written as 𝑻𝒆𝒎 =

𝒑𝑳𝒎

𝑳𝒔(𝑰𝒓𝒒 𝒔𝒅) (13)

In the field of wind energy production, these are average and high power machines; ignoring the stator resistance and taking the constant stator flux, we can write

𝑽𝒔𝒅 = 𝟎

𝑽𝒔𝒒 = 𝒘𝒔𝜱𝒔𝒅 = 𝑽𝒔

Taking into account the simplifications, the expression of the stator powers are

𝑷𝒔 = −𝑽𝒔𝑳𝒎

𝑳𝒔𝑰𝒓𝒒 (14)

𝑸𝒔 = −𝑽𝒔𝑳𝒎

𝑳𝒔𝑰𝒓𝒅 +

𝑽𝒔𝟐

𝑳𝒔𝒘𝒔 (15)

In order to properly control the machine, we then need to establish the relationship between the currents and the rotor voltages that will be applied to the machine [14] [15].

𝑽𝒓𝒒 = 𝑹𝒓𝑰𝒓𝒒 + 𝝈𝑳𝒓𝒅𝑰𝒓𝒒

𝒅𝒕− 𝒈𝒘𝒔𝝈𝑳𝒓𝑰𝒓𝒅

+ 𝒈𝑳𝒎𝑽𝒔

𝑳𝒔 (16)

𝑽𝒓𝒅 = 𝑹𝒓𝑰𝒓𝒅 + 𝝈𝑳𝒓𝒅𝑰𝒓𝒅

𝒅𝒕− 𝒈𝒘𝒔𝝈𝑳𝒓𝑰𝒓𝒒 (17)

4.2 Application of first order sliding mode

control The advantages of sliding mode control are significant and multiple: high precision, good stability, simplicity, invariance, and robustness. This allows it to be particularly suited to the system with a vague template [15]. In sliding mode theory, the structure control law has two parts:

A continuous part presents the dynamic of system during the sliding mode.

A discontinuous part (switching control) presents the dynamic during the convergence mode.

The design of the command can be performed in three major steps, dependent on each other [15] [16].

Choose the sliding mode surface. Find the existing conditions of sliding mode

control. Determine the law of control.

4.2.1 Control of active power

We apply the expression of Slotine [9] to find the active power sliding surface: 𝑺(𝑷𝒔) = 𝑷𝒔𝒓𝒆𝒇 − 𝑷𝒔 (18)

The derivative of the sliding surface is ��(𝑷𝒔) = ��𝒓𝒆𝒇 − ��𝒔 (19)

Replacing the expression of active power described by equation (14), we obtain ��(𝑷𝒔) = ��𝒔𝒓𝒆𝒇 + 𝑽𝒔

𝑳𝒎

𝑳𝒔��𝒓𝒒 (20)

We conclude the expression of the current derivative from equation (16) by compensating the disturbance terms and ignoring the coupling terms between the control axis due to small slip value (g=0.03). We obtain, ��(𝑷𝒔) = ��𝒔𝒓𝒆𝒇 − 𝑽𝒔

𝑳𝒎

𝝈𝑳𝒓𝑳𝒔(𝑽𝒓𝒒 − 𝑹𝒓𝑰𝒓𝒒) (21)



E-ISSN: 2224-2678 74 Volume 17, 2018

The control algorithm appears in the following equation: ��(𝑷𝒔) = ��𝒔𝒓𝒆𝒇 − 𝑽𝒔

𝑳𝒎

𝝈𝑳𝒓𝑳𝒔((𝑽𝒓𝒒

𝒆𝒒+ 𝑽𝒓𝒒

𝒏 ) − 𝑹𝒓𝑰𝒓𝒒) (22)

During sliding mode and permanent scheme, the derivative of the surface is zero (because the error is equal to zero), plus attractive control 𝑉𝑟𝑞

𝑛 (which does not exist in this mode). The equivalent control is given by 𝑽𝒓𝒒

𝒆𝒒= −

𝝈𝑳𝒓𝑳𝒔

𝑽𝒔.𝑳𝒎��𝒔𝒓𝒆𝒇 + 𝑹𝒓𝑰𝒓𝒒 (23)

This command ensures that the trajectory of the controlled quantity remains on its sliding surface. However, this command does not ensure control when the dynamic behavior is not found on the sliding surface. Therefore, it is necessary to add a new condition to ensure that the system state tends towards the surface and reaches it. This new condition is the condition of attractiveness [17] [18] [19]. The sliding mode exists if and only if the following condition is true: 𝑺(𝑷𝒔)��(𝑷𝒔) < 𝟎 (24)

During the convergence mode, by replacing the term 𝑽𝒓𝒒

𝒆𝒒 in equation (22), we obtain a new expression of the derivative of the surface, as follows: ��(𝑷𝒔) = 𝑽𝒔

𝑳𝒎

𝝈𝑳𝒓𝑳𝒔𝑽𝒓𝒒

𝒏 (25)

According to the arrival approach [8] (arrival law with a speed of constant arrival), the switching term is given by 𝑽𝒓𝒒

𝒏 = −𝑸𝒔𝒊𝒈𝒏(𝑺(𝑷𝒔)) (26)

4.2.1 Control of reactive power

To control the reactive power, we adapt the same procedure applied in active power control by replacing the symbols 𝑷𝒔, 𝑷𝒔𝒓𝒆𝒇, 𝑰𝒓𝒒, and 𝑽𝒓𝒒, respectively, with 𝑸𝒔, 𝑸𝒔𝒓𝒆𝒇, 𝑰𝒓𝒅, and 𝑽𝒓𝒅. 5 Simulation Results The strategy of field-oriented control with a first order sliding mode regulator of stator active and reactive power with power electronic interface using PWM technique has been implemented in the Matlab Simulink* environment to evaluate and test the

control of the complete system, as shown in Fig.7. The wind turbine system of 1.5MW is based on DFIG and Table 1 shows the main characteristics of the wind power generation system.

Figure 7. Diagram of the proposed control for

Wind turbine system

Designation and symbols Values

Nominal power. 𝑃𝑛 1.5 MW Stator resistance. 𝑅𝑠 0.012 Ω Rotor resistance. 𝑅𝑟 0.021 Ω Stator inductance. 𝐿𝑠 0.0137 H Rotor inductance. 𝐿𝑟 0.0136 H

Magnetizing inductance. 𝐿𝑚 0.013 H Turbine radius. r 35.25 m

Blades numbers. - 3 Gain multiplier. G 90

Total inertia. J 1000 Kgm2 Table 1. Parameters of Wind turbine system

Fig. 8 shows the wind speed variation. The wind profile begins with a wind speed equal to the nominal values (15.5m/s). This value is increased after one second to apply the high speed wind values (20 m/s) on the aero generator system to test the PITCH system. At 1.5s, we apply wind speed values lower than the nominal to examine the MPPT algorithm.

Figure 8. Speed of the wind

* Simulink, Matlab de © 1994-2015 The Math

Works, Inc.

0 0.5 1 1.5 2 2.5 310

12

14

16

18

20

22

t[Sec]

Win

d s

pe

ed

[m

/s]

V = Vn

V > Vn

V < Vn

V = Vn



E-ISSN: 2224-2678 75 Volume 17, 2018

The main objective of mechanical parts control of a wind turbine system is to extract the maximum existing wind power using the MPPT algorithm when wind speed values are lower than the nominal and to protect the turbine by decreasing the yield when the speed wind is higher than the nominal value using the PITCH system. Fig. 9 shows the mechanical speed dynamics of the DFIG shaft when the variable amplitude steps of stator active and reactive power were applied.

Figure 9. Variations of the wind speed and the

mechanical rotation speed without PITCH

system and MPPT algorithm peed of the wind The same reference levels are applied to the DFIG input, but we integrate the mechanical control technique without rotational speed control (Pitch system and MPPT algorithm). Note that mechanical speed variation is controlled, as shown in Fig. 10. When the wind reaches the nominal speed, the mechanical speed reaches its nominal value (163 rad/s). If the wind speed increases, the PITCH system decreases the yield of the wind turbine to keep the nominal value and protect the wind (1−1.5s). When the wind speed decreases, the MPPT algorithm is applied to extract the maximum wind power (1.5−2s).

Figure 10. Wind speed and mechanical speed

with MPPT and PITCH

Fig. 11 shows the simulation results for field-oriented control with a first order sliding mode regulator of the DFIG associated with a static converter integrated in

a wind system when the reference steps of power were applied.

Figure 11. Active and reactive power

The wind generation system must extract the maximum wind power when the speed is lower than the nominal values [2]; this power corresponds to the active power generated by DFIG. The reactive power must be zero to ensure a unity power factor on the stator side. Fig.12 presents the estimated active power from the MPPT power (electromagnetic torque controlled by the MPPT algorithm and PITCH system) and reactive power.

We observe that the system extracts the nominal values of active power (1.5MW, maximum power can be generated by this system) regardless of the existing wind power. This power limitation is used to avoid a possible over-speed of the rotor and to protect the electrical system (the generator). When the wind speed decreases, the MPPT algorithm applies the maximized power extracted from the wind (1.5−2s).

Figure 12. Power active estimated and reactive

power zero

Fig. 13 presents the simulation results of MPPT and PITCH electromagnetic torque generated from the mechanical part as a reference (red), the electromagnetic torque developed by DFIG (blue), and the stator currents.

0 0.5 1 1.5 2 2.5 310

15

20

t [Sec]

Win

d s

pe

ed

[m

/s]

0 0.5 1 1.5 2 2.5 30

100

200

300

t [Sec]

Me

ch

an

ica

l sp

ee

d [ra

d/s

]

V = Vn

V > Vn V < V

n

over speed

0 0.5 1 1.5 2 2.5 310

15

20

t [Sec]

Win

d s

pe

ed

[m

/s]

0 0.5 1 1.5 2 2.5 30

100

200

t [Sec]

Me

ch

an

ica

l sp

ee

d [ra

d/s

]

V = Vn

V > Vn V < V

n

PITCH System MPPT Algorithm

0 0.5 1 1.5 2 2.5 3-2

0

2x 10

5

Active

po

we

re [W

]

t [Sec]

0 0.5 1 1.5 2 2.5 3-2

0

2x 10

5

Re

active

po

we

r [V

Ar]

t [Sec]

real

reference

real

reference

0 0.5 1 1.5 2 2.5 3-5

0

5

10

15x 10

5

Active

po

we

re [W

]

t [Sec]

0 0.5 1 1.5 2 2.5 3-2

0

2x 10

5

Re

active

po

we

r [V

Ar]

t [Sec]

Electric active power

MPPT power

real

reference

V = Vn et

= n

V < Vn et

< n

V > Vn et

= n

PITCH Syetem M PPT Algorithm



E-ISSN: 2224-2678 76 Volume 17, 2018

Figure 13. Electromagnetic torque and stator

currents of the overall system

Figure 14. Variation of the relative speed and the

angle

Fig. 14 presents the variation of the relative speed and the blade angle. Note that the wind speed is equal to the nominal value, the relative speed is equal to the optimum value 𝝀𝒐𝒑𝒕 , and β is equal to 𝛃𝒐𝒑𝒕 . If the wind speed increases, the relative ration speed decreases and the angle timing increases (PITCH system). If the wind speed takes a value lower than the nominal value, then the relative velocity increases and returns to its optimal value. 6 Conclusions Regulation by means of a first order sliding mode regulator shows a good performance even in the presence of changes in instructions. Decoupling ensures stability and convergence. In addition, this setting presents a very simple robust control algorithm.

The interest of the PITCH system in the protection of mechanical part and electric part (generator) is achieved when the rate of wind speed is exceeded.

The interest of the MPPT algorithm in the extraction of the maximum wind power is acquired when the speed of the wind is decreased such that a high power yield of the generation system is implied.

References:

[1] Belmokhtar K., Doumbia L. and Agbossou K. “Modelling and control of a wind energy system

down to supply power to the DFIM grid” (French) , 4th International Conference on Electrical Engineering, CIGE’10. University of Bechar, Algeria. 03-04 November 2010. [2] Beltran B. “Contribution to the robust control of

wind turbines based on DFIG: the classic sliding

mode and higher order sliding mode” (French). PhD thesis, University of Western Brittany, 2010. [3] Bianchi FD., De Battista H. and Mantz RJ. “Wind

turbine control systems: principles, modelling and

gain scheduling design”. Springer, London. 2007. [4] Beltran B., Benbouzid MEH. and Ahmed-Ali T. “High-order sliding mode control of a DFIG-based

wind turbine for power maximization and grid fault

tolerance”. In: Proceedings of IEEE IEMDC. Miami, USA. 2009. [5] Bouhedda A. “Contribution to the study of a wind

turbine control systems” (French). PhD thesis, University of Mouloud Mammeri. Tizi-Ouzou, Algeria, 2011. [6] Poitiers F. “Study and Control of asynchronous

generators for the use in wind power” (French). PhD thesis, Polytechnic School of Nantes, France, 2003. [7] Cardenas R., Pena R., Alepuz S. and Asher G. “Overview of control systems for the operation of

DFIGs in wind energy aplications”. IEEE Trans Ind Electron, pp 2776-2798. 2013 [8] Itaso Martinez M., Susperregui A., Tapia G. and Camblong H. “Sliding-Mode Control for a DFIG-

based Wind Turbine under Unbalanced Voltage”. Preprints of the 18th IFAC Word Congress Milano (Italy). 2011. [9] Hamdi N. “Modeling and control of wind power

generators” (French). Magister thesis in Electrical Engineering, University of Mentouri, Algeria, 2008. [10] Djeriri L. “Vector control of a DFIM integrated

in wind system” (French). PhD thesis, University of Djilali Liabes of Sidi-Bel-Abbes, Algeria, 2008. [11] Gaillard A. “Wind turbine system based on a

DFIM: Contribution to the study of the quality of

electric power and continuity Service” (French). PhD thesis, University of Henri Poincare Nancy I, France, 2010. [12] Djilali L. “FOC Comparative study of a wind

power system a basis of DFIG with and without

control of the mechanical part” (French). 2nd International Conference on electronics, electrical and automatic. Oran, Algeria, 26-27 November 2013. [13] Mejía D., Torres I., and Diaz J. “MPPT control

systems used in sunflower for maximum efficiency”. Revista Colombiana de Tecnologías de Avanzada, pp. 118-123. ISSN: 1692-7257-Vol. 1.2016.

0 0.5 1 1.5 2 2.5 3-5000

0

5000

sta

tor

cu

rre

nts

[A

]

t [Sec]

0 0.5 1 1.5 2 2.5 3-5000

0

5000

10000

ele

ctr

om

ag

ne

tic to

rqu

e

[N

.m]

t [Sec]

reel

reference

V = Vn et

= n

V > Vn et

= n

V < Vn et <

n

PITCH system

M PPT algorithm

0 1 2 33

4

5

6

7

8

9

10

t [Sec]

0 1 2 30

1

2

3

4

5

6

t [Sec]

0

V > Vn

V < Vn

V = Vn

V = Vn

V > Vn

V < Vn

V = Vn V = V

n

(a) (b)



E-ISSN: 2224-2678 77 Volume 17, 2018

[14] Giménez M. “Develop advance in new

strategies for electric machine control base on DTC

(Direct Torque Control)”. Revista Colombiana de Tecnologías de Avanzada, pp. 27-36. ISSN: 1692-7257-Vol. 1.2007. [15] Bachtarzi A. “Control of variable structure

systems Applications to a steam generator” (French), PhD thesis, University of Constantine, Algeria. 2011. [16] Beltran B., Benbouzid MEH. and Ahmed-Ali T. “DFIG-based wind turbine robust control using

high-order sliding modes and a high gain observer”. HAL archives-ouvertes. International Review on Modelling and Simulations, pp. 1148-1155. 2011. [17] Bermeo W., Souza Jr A., Fernandes T., Honórico D., Nogueira L., and Barreto L. “Sliding

model control applied in current loop for a DSP-

based position control applied to squirrel-cage

induction motor”. Revista Colombiana de Tecnologías de Avanzada, pp. 26-32. ISSN: 1692-7257-Vol. 1.2016. [18] Utkin V., Guldner J. and Shi J. “Sliding Mode

Control in Electromechanical Systems”. Taylor&Francis, London, UK. 1999. [19] Álvarez R., García M., and Torregoza A. “Photovoltaic module-integrated converter (mic)

based on semi-z-source inverters (szsi): a new sliding

mode control scheme”. Revista Colombiana de Tecnologías de Avanzada, pp. 61-66. ISSN: 1692-7257-Vol. 1.2014.



E-ISSN: 2224-2678 78 Volume 17, 2018

Download - First Sliding Mode Control of a DFIG-based Wind Turbine ...source of energy. The performance of a wind turbine depends essentially on three parameters: the power of the wind, the turbine

Top Related